Fuzzy Time Series Based on K-means and Particle Swarm Optimization Algorithm

The unequal-sized intervals have important influence on the prediction of the fuzzy time series. With the development of intelligent optimization algorithm increasingly, the theory of K-means based on particle swarm algorithm is put forward to determine the length of interval, making full use the strong local searching ability of K-means and the good global searching ability of particle swarm optimization, overcoming the disadvantage that the result relies on the initial value of K-means too much, and getting the optimal length of interval. Finally, the 22 years of enrollment of Alabama is used to verify the feasibility of the model by comparing and analyzing the mean square error of K-means model and the KPSO method.